eval_gsm8k_chat.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import re\n",
    "import math\n",
    "import random\n",
    "from typing import List\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import argparse\n",
    "import torch\n",
    "import datasets\n",
    "from tqdm import tqdm\n",
    "from transformers.trainer_utils import set_seed\n",
    "from transformers import AutoModelForCausalLM, AutoTokenizer\n",
    "from transformers import BitsAndBytesConfig\n",
    "from transformers.generation import GenerationConfig\n",
    "from datasets import load_dataset, load_from_disk\n",
    "\n",
    "\n",
    "class args:\n",
    "    checkpoint_path = '/gemini/code/lamma3_eval/lamma3_model/8B_instruct'\n",
    "    eval_data_path = '/gemini/code/lamma3_eval/eval_data/gsm8k'\n",
    "    save_result_dir = \"/gemini/code/lamma3_eval/eval_result/gsm8k_chat\"\n",
    "    # choices = [\"A\", \"B\", \"C\", \"D\"]\n",
    "    debug = False\n",
    "    overwrite = False\n",
    "    batch_size = 6"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# dataset = datasets.load_dataset(\"gsm8k\",'main')\n",
    "# dataset.save_to_disk(args.eval_data_path)\n",
    "\n",
    "dataset = load_from_disk(args.eval_data_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'question': \"Janet’s ducks lay 16 eggs per day. She eats three for breakfast every morning and bakes muffins for her friends every day with four. She sells the remainder at the farmers' market daily for $2 per fresh duck egg. How much in dollars does she make every day at the farmers' market?\",\n",
       " 'answer': 'Janet sells 16 - 3 - 4 = <<16-3-4=9>>9 duck eggs a day.\\nShe makes 9 * 2 = $<<9*2=18>>18 every day at the farmer’s market.\\n#### 18'}"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset['test'][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_models_tokenizer():\n",
    "    tokenizer = AutoTokenizer.from_pretrained(\n",
    "        args.checkpoint_path,\n",
    "        padding_side='left'\n",
    "    )\n",
    "\n",
    "    quantization_config = BitsAndBytesConfig(\n",
    "        load_in_4bit=True,\n",
    "        bnb_4bit_compute_dtype=torch.bfloat16,\n",
    "        bnb_4bit_use_double_quant=True,\n",
    "        bnb_4bit_quant_type=\"nf4\")\n",
    "    \n",
    "    model = AutoModelForCausalLM.from_pretrained(\n",
    "        args.checkpoint_path,\n",
    "        device_map=\"auto\",\n",
    "        # quantization_config=quantization_config\n",
    "        # torch_dtype=torch.bfloat16\n",
    "    ).eval()\n",
    "    model.generation_config = GenerationConfig.from_pretrained(\n",
    "        args.checkpoint_path\n",
    "    )\n",
    "    model.generation_config.do_sample = False  # use greedy decoding\n",
    "    model.generation_config.repetition_penalty = 1.0  # disable repetition penalty\n",
    "    return model, tokenizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n",
      "Loading checkpoint shards: 100%|██████████| 4/4 [00:58<00:00, 14.65s/it]\n"
     ]
    }
   ],
   "source": [
    "model, tokenizer = load_models_tokenizer()\n",
    "tokenizer.pad_token_id = tokenizer.eos_token_id"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "fewshot_prompt = open(\"gsm8k_prompt.txt\").read()\n",
    "prompt = \"Here is 8 examples, please answer the last question in this format.\\n\\n\"\n",
    "start_prompt = \"\\nPay attention! You need give your answer at the end in the form of 'The answer is '\\nHere is my question:\\n\"\n",
    "\n",
    "\n",
    "def batch_process(func, *args):\n",
    "    '''\n",
    "    args 负责接受 一个或多个 batch\n",
    "    '''\n",
    "    # print(f'args len: {len(args)}')\n",
    "    \n",
    "    text_ls = []\n",
    "    for sample in zip(*args):\n",
    "        text_ls.append(func(*sample))\n",
    "    return text_ls\n",
    "\n",
    "\n",
    "def doc_to_text_item(doc):\n",
    "    return (\n",
    "        prompt\n",
    "        + fewshot_prompt\n",
    "        + start_prompt\n",
    "        + \"\\nQuestion: \"\n",
    "        + doc\n",
    "        + \"\\nLet's think step by step\\n\"\n",
    "    )\n",
    "\n",
    "def doc_to_text(doc):\n",
    "    return batch_process(doc_to_text_item, doc)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def clear_output_item(text):  # 可能需要改进，可以把raw_text长度传进来，直接截断，然后提取第一个回答。这里相当 取了最后一个回答\n",
    "    left_extract = \"<|start_header_id|>assistant<|end_header_id|>\"\n",
    "    st = text.rfind(left_extract) + len(left_extract)\n",
    "    output = text[st:]\n",
    "\n",
    "    stop_words = [\"<|end_of_text|>\", \"<|eot_id|>\"]\n",
    "    for sw in stop_words:\n",
    "        output = output.replace(sw, \"\").strip()\n",
    "    \n",
    "    if output == \"\": print(text)\n",
    "    return output\n",
    "\n",
    "def clear_output(text):\n",
    "    return batch_process(clear_output_item, text)\n",
    "\n",
    "\n",
    "def generate_sample(model, tokenizer, input_txt):\n",
    "    chat_template = [[{'content': t, 'role': 'user'}] for t in input_txt]\n",
    "    input_txt = tokenizer.apply_chat_template(chat_template, tokenize=False, add_generation_prompt=True)\n",
    "    input_ids = tokenizer(input_txt, padding=True, return_tensors=\"pt\").to(model.device)\n",
    "\n",
    "    # print(input_ids['input_ids'])\n",
    "    # print(input_ids['attention_mask'])\n",
    "\n",
    "    outputs_id = model.generate(**input_ids, max_new_tokens = 512, eos_token_id = 128009, pad_token_id = tokenizer.pad_token_id,\n",
    "                                repetition_penalty = 1.2, do_sample = False, temperature = 1.0, top_p = 1.0)\n",
    "    # print(outputs)\n",
    "    outputs = tokenizer.batch_decode(outputs_id, skip_special_tokens=False)\n",
    "    answer = clear_output(outputs)\n",
    "\n",
    "    return answer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def extract_answer_item(s):\n",
    "    _PAT_LAST_DIGIT = re.compile(\n",
    "        r\"([+-])?(?=([0-9]|\\.[0-9]))(0|([1-9](\\d{0,2}(,\\d{3})*)|\\d*))?(\\.\\d*)?(?=\\D|$)\"\n",
    "    )\n",
    "    match = list(_PAT_LAST_DIGIT.finditer(s))\n",
    "    if match:\n",
    "        last_digit = match[-1].group().replace(\",\", \"\").replace(\"+\", \"\").strip()\n",
    "        # print(f\"The last digit in {s} is {last_digit}\")\n",
    "    else:\n",
    "        last_digit = None\n",
    "        # print(f\"No digits found in {s!r}\", flush=True)\n",
    "    return last_digit\n",
    "\n",
    "def extract_answer(s):\n",
    "    return batch_process(extract_answer_item, s)\n",
    "\n",
    "\n",
    "def is_correct_item(completion, answer):\n",
    "    predict = extract_answer_item(completion)\n",
    "    gold = extract_answer_item(answer)\n",
    "\n",
    "    assert gold is not None, \"No ground truth answer found in the document.\"\n",
    "\n",
    "    if predict is None:\n",
    "        return False\n",
    "    try:\n",
    "        return math.isclose(eval(gold), eval(predict), rel_tol=0, abs_tol=1e-4)\n",
    "    except:\n",
    "        print(\n",
    "            f\"cannot compare two numbers: answer={gold}, pred={predict}\", flush=True\n",
    "        )\n",
    "        return False\n",
    "    \n",
    "\n",
    "def is_correct(completion, answer):\n",
    "    return batch_process(is_correct_item, completion, answer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "@torch.no_grad()\n",
    "def main():\n",
    "    all_sz = len(dataset['test'])\n",
    "    # test = dataset['test'].select(random.sample(range(all_sz),150))\n",
    "    test = dataset['test']\n",
    "    acc_ls = []\n",
    "\n",
    "    # 看有无文件，有的话就不重复做了\n",
    "    result_path = os.path.join(args.save_result_dir, f\"result.csv\")\n",
    "    if not args.overwrite and os.path.exists(result_path):\n",
    "        print(f\"{result_path} existed, skip!\")\n",
    "        for (_, resultrow) in pd.read_csv(result_path).iterrows():\n",
    "            # pred = extract_answer(resultrow['model_response'], datarow)\n",
    "            acc = resultrow[\"ACC\"]\n",
    "            acc_ls.append(acc)\n",
    "\n",
    "    else:\n",
    "        question_ls = []\n",
    "        answer_ls = []\n",
    "        response_ls = []\n",
    "\n",
    "        for i in tqdm(range(0, len(test), args.batch_size)):\n",
    "            batch = test.select(range(i, min(i+args.batch_size, len(test))))\n",
    "\n",
    "            context = doc_to_text(batch[\"question\"])\n",
    "            question_ls.extend(batch[\"question\"])\n",
    "            # print(context)\n",
    "            completion = generate_sample(model, tokenizer, context)\n",
    "            response_ls.extend(completion)\n",
    "            \n",
    "            acc = is_correct(completion, batch[\"answer\"])\n",
    "            answer_ls.extend(batch[\"answer\"])\n",
    "            acc_ls.extend(acc)\n",
    "\n",
    "        # 存入文件\n",
    "        output_df = pd.DataFrame(\n",
    "            {\"model_question\": question_ls,\n",
    "             \"standard_answer\": answer_ls,\n",
    "             \"model_response\": response_ls, \n",
    "             \"ACC\": acc_ls}\n",
    "        )\n",
    "        os.makedirs(args.save_result_dir, exist_ok=True)\n",
    "        output_df.to_csv(\n",
    "            result_path,\n",
    "            encoding=\"utf-8\",\n",
    "            index=False\n",
    "        )\n",
    "\n",
    "\n",
    "    print(\"AVERAGE ACC:%.2f \" % (sum(acc_ls) / len(acc_ls) * 100))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  1%|          | 1/165 [01:19<3:37:55, 79.73s/it]\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[12], line 2\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;18m__name__\u001b[39m \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m__main__\u001b[39m\u001b[38;5;124m\"\u001b[39m:\n\u001b[0;32m----> 2\u001b[0m     \u001b[43mmain\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/miniconda3/lib/python3.10/site-packages/torch/utils/_contextlib.py:115\u001b[0m, in \u001b[0;36mcontext_decorator.<locals>.decorate_context\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    112\u001b[0m \u001b[38;5;129m@functools\u001b[39m\u001b[38;5;241m.\u001b[39mwraps(func)\n\u001b[1;32m    113\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mdecorate_context\u001b[39m(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs):\n\u001b[1;32m    114\u001b[0m     \u001b[38;5;28;01mwith\u001b[39;00m ctx_factory():\n\u001b[0;32m--> 115\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfunc\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
      "Cell \u001b[0;32mIn[10], line 28\u001b[0m, in \u001b[0;36mmain\u001b[0;34m()\u001b[0m\n\u001b[1;32m     26\u001b[0m question_ls\u001b[38;5;241m.\u001b[39mextend(batch[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mquestion\u001b[39m\u001b[38;5;124m\"\u001b[39m])\n\u001b[1;32m     27\u001b[0m \u001b[38;5;66;03m# print(context)\u001b[39;00m\n\u001b[0;32m---> 28\u001b[0m completion \u001b[38;5;241m=\u001b[39m \u001b[43mgenerate_sample\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmodel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtokenizer\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcontext\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     29\u001b[0m response_ls\u001b[38;5;241m.\u001b[39mextend(completion)\n\u001b[1;32m     31\u001b[0m acc \u001b[38;5;241m=\u001b[39m is_correct(completion, batch[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124manswer\u001b[39m\u001b[38;5;124m\"\u001b[39m])\n",
      "Cell \u001b[0;32mIn[8], line 25\u001b[0m, in \u001b[0;36mgenerate_sample\u001b[0;34m(model, tokenizer, input_txt)\u001b[0m\n\u001b[1;32m     20\u001b[0m input_ids \u001b[38;5;241m=\u001b[39m tokenizer(input_txt, padding\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m, return_tensors\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mpt\u001b[39m\u001b[38;5;124m\"\u001b[39m)\u001b[38;5;241m.\u001b[39mto(model\u001b[38;5;241m.\u001b[39mdevice)\n\u001b[1;32m     22\u001b[0m \u001b[38;5;66;03m# print(input_ids['input_ids'])\u001b[39;00m\n\u001b[1;32m     23\u001b[0m \u001b[38;5;66;03m# print(input_ids['attention_mask'])\u001b[39;00m\n\u001b[0;32m---> 25\u001b[0m outputs_id \u001b[38;5;241m=\u001b[39m \u001b[43mmodel\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mgenerate\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43minput_ids\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmax_new_tokens\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m \u001b[49m\u001b[38;5;241;43m512\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43meos_token_id\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m \u001b[49m\u001b[38;5;241;43m128009\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mpad_token_id\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m \u001b[49m\u001b[43mtokenizer\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpad_token_id\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m     26\u001b[0m \u001b[43m                            \u001b[49m\u001b[43mrepetition_penalty\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m \u001b[49m\u001b[38;5;241;43m1.2\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdo_sample\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mFalse\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtemperature\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m \u001b[49m\u001b[38;5;241;43m1.0\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtop_p\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m \u001b[49m\u001b[38;5;241;43m1.0\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[1;32m     27\u001b[0m \u001b[38;5;66;03m# print(outputs)\u001b[39;00m\n\u001b[1;32m     28\u001b[0m outputs \u001b[38;5;241m=\u001b[39m tokenizer\u001b[38;5;241m.\u001b[39mbatch_decode(outputs_id, skip_special_tokens\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mFalse\u001b[39;00m)\n",
      "File \u001b[0;32m~/miniconda3/lib/python3.10/site-packages/torch/utils/_contextlib.py:115\u001b[0m, in \u001b[0;36mcontext_decorator.<locals>.decorate_context\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    112\u001b[0m \u001b[38;5;129m@functools\u001b[39m\u001b[38;5;241m.\u001b[39mwraps(func)\n\u001b[1;32m    113\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mdecorate_context\u001b[39m(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs):\n\u001b[1;32m    114\u001b[0m     \u001b[38;5;28;01mwith\u001b[39;00m ctx_factory():\n\u001b[0;32m--> 115\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfunc\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/miniconda3/lib/python3.10/site-packages/transformers/generation/utils.py:1576\u001b[0m, in \u001b[0;36mGenerationMixin.generate\u001b[0;34m(self, inputs, generation_config, logits_processor, stopping_criteria, prefix_allowed_tokens_fn, synced_gpus, assistant_model, streamer, negative_prompt_ids, negative_prompt_attention_mask, **kwargs)\u001b[0m\n\u001b[1;32m   1559\u001b[0m     result \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_assisted_decoding(\n\u001b[1;32m   1560\u001b[0m         input_ids,\n\u001b[1;32m   1561\u001b[0m         candidate_generator\u001b[38;5;241m=\u001b[39mcandidate_generator,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m   1572\u001b[0m         \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mmodel_kwargs,\n\u001b[1;32m   1573\u001b[0m     )\n\u001b[1;32m   1574\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m generation_mode \u001b[38;5;241m==\u001b[39m GenerationMode\u001b[38;5;241m.\u001b[39mGREEDY_SEARCH:\n\u001b[1;32m   1575\u001b[0m     \u001b[38;5;66;03m# 11. run greedy search\u001b[39;00m\n\u001b[0;32m-> 1576\u001b[0m     result \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_greedy_search\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1577\u001b[0m \u001b[43m        \u001b[49m\u001b[43minput_ids\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1578\u001b[0m \u001b[43m        \u001b[49m\u001b[43mlogits_processor\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mprepared_logits_processor\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1579\u001b[0m \u001b[43m        \u001b[49m\u001b[43mstopping_criteria\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mprepared_stopping_criteria\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1580\u001b[0m \u001b[43m        \u001b[49m\u001b[43mpad_token_id\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mgeneration_config\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpad_token_id\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1581\u001b[0m \u001b[43m        \u001b[49m\u001b[43moutput_scores\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mgeneration_config\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43moutput_scores\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1582\u001b[0m \u001b[43m        \u001b[49m\u001b[43moutput_logits\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mgeneration_config\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43moutput_logits\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1583\u001b[0m \u001b[43m        \u001b[49m\u001b[43mreturn_dict_in_generate\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mgeneration_config\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mreturn_dict_in_generate\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1584\u001b[0m \u001b[43m        \u001b[49m\u001b[43msynced_gpus\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43msynced_gpus\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1585\u001b[0m \u001b[43m        \u001b[49m\u001b[43mstreamer\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mstreamer\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1586\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mmodel_kwargs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1587\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1589\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m generation_mode \u001b[38;5;241m==\u001b[39m GenerationMode\u001b[38;5;241m.\u001b[39mCONTRASTIVE_SEARCH:\n\u001b[1;32m   1590\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m model_kwargs[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124muse_cache\u001b[39m\u001b[38;5;124m\"\u001b[39m]:\n",
      "File \u001b[0;32m~/miniconda3/lib/python3.10/site-packages/transformers/generation/utils.py:2494\u001b[0m, in \u001b[0;36mGenerationMixin._greedy_search\u001b[0;34m(self, input_ids, logits_processor, stopping_criteria, max_length, pad_token_id, eos_token_id, output_attentions, output_hidden_states, output_scores, output_logits, return_dict_in_generate, synced_gpus, streamer, **model_kwargs)\u001b[0m\n\u001b[1;32m   2491\u001b[0m model_inputs \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mprepare_inputs_for_generation(input_ids, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mmodel_kwargs)\n\u001b[1;32m   2493\u001b[0m \u001b[38;5;66;03m# forward pass to get next token\u001b[39;00m\n\u001b[0;32m-> 2494\u001b[0m outputs \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[43m(\u001b[49m\n\u001b[1;32m   2495\u001b[0m \u001b[43m    \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mmodel_inputs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   2496\u001b[0m \u001b[43m    \u001b[49m\u001b[43mreturn_dict\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[1;32m   2497\u001b[0m \u001b[43m    \u001b[49m\u001b[43moutput_attentions\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43moutput_attentions\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   2498\u001b[0m \u001b[43m    \u001b[49m\u001b[43moutput_hidden_states\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43moutput_hidden_states\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   2499\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   2501\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m synced_gpus \u001b[38;5;129;01mand\u001b[39;00m this_peer_finished:\n\u001b[1;32m   2502\u001b[0m     \u001b[38;5;28;01mcontinue\u001b[39;00m  \u001b[38;5;66;03m# don't waste resources running the code we don't need\u001b[39;00m\n",
      "File \u001b[0;32m~/miniconda3/lib/python3.10/site-packages/torch/nn/modules/module.py:1501\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1496\u001b[0m \u001b[38;5;66;03m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1497\u001b[0m \u001b[38;5;66;03m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1498\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m   1499\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1500\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mforward_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1502\u001b[0m \u001b[38;5;66;03m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m   1503\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[38;5;241m=\u001b[39m [], []\n",
      "File \u001b[0;32m~/miniconda3/lib/python3.10/site-packages/transformers/models/llama/modeling_llama.py:1210\u001b[0m, in \u001b[0;36mLlamaForCausalLM.forward\u001b[0;34m(self, input_ids, attention_mask, position_ids, past_key_values, inputs_embeds, labels, use_cache, output_attentions, output_hidden_states, return_dict, cache_position)\u001b[0m\n\u001b[1;32m   1207\u001b[0m return_dict \u001b[38;5;241m=\u001b[39m return_dict \u001b[38;5;28;01mif\u001b[39;00m return_dict \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mconfig\u001b[38;5;241m.\u001b[39muse_return_dict\n\u001b[1;32m   1209\u001b[0m \u001b[38;5;66;03m# decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn)\u001b[39;00m\n\u001b[0;32m-> 1210\u001b[0m outputs \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mmodel\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1211\u001b[0m \u001b[43m    \u001b[49m\u001b[43minput_ids\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43minput_ids\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1212\u001b[0m \u001b[43m    \u001b[49m\u001b[43mattention_mask\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mattention_mask\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1213\u001b[0m \u001b[43m    \u001b[49m\u001b[43mposition_ids\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mposition_ids\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1214\u001b[0m \u001b[43m    \u001b[49m\u001b[43mpast_key_values\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mpast_key_values\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1215\u001b[0m \u001b[43m    \u001b[49m\u001b[43minputs_embeds\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43minputs_embeds\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1216\u001b[0m \u001b[43m    \u001b[49m\u001b[43muse_cache\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43muse_cache\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1217\u001b[0m \u001b[43m    \u001b[49m\u001b[43moutput_attentions\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43moutput_attentions\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1218\u001b[0m \u001b[43m    \u001b[49m\u001b[43moutput_hidden_states\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43moutput_hidden_states\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1219\u001b[0m \u001b[43m    \u001b[49m\u001b[43mreturn_dict\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mreturn_dict\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1220\u001b[0m \u001b[43m    \u001b[49m\u001b[43mcache_position\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcache_position\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1221\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1223\u001b[0m hidden_states \u001b[38;5;241m=\u001b[39m outputs[\u001b[38;5;241m0\u001b[39m]\n\u001b[1;32m   1224\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mconfig\u001b[38;5;241m.\u001b[39mpretraining_tp \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m1\u001b[39m:\n",
      "File \u001b[0;32m~/miniconda3/lib/python3.10/site-packages/torch/nn/modules/module.py:1501\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1496\u001b[0m \u001b[38;5;66;03m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1497\u001b[0m \u001b[38;5;66;03m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1498\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m   1499\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1500\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mforward_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1502\u001b[0m \u001b[38;5;66;03m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m   1503\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[38;5;241m=\u001b[39m [], []\n",
      "File \u001b[0;32m~/miniconda3/lib/python3.10/site-packages/transformers/models/llama/modeling_llama.py:1018\u001b[0m, in \u001b[0;36mLlamaModel.forward\u001b[0;34m(self, input_ids, attention_mask, position_ids, past_key_values, inputs_embeds, use_cache, output_attentions, output_hidden_states, return_dict, cache_position)\u001b[0m\n\u001b[1;32m   1007\u001b[0m     layer_outputs \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_gradient_checkpointing_func(\n\u001b[1;32m   1008\u001b[0m         decoder_layer\u001b[38;5;241m.\u001b[39m\u001b[38;5;21m__call__\u001b[39m,\n\u001b[1;32m   1009\u001b[0m         hidden_states,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m   1015\u001b[0m         cache_position,\n\u001b[1;32m   1016\u001b[0m     )\n\u001b[1;32m   1017\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m-> 1018\u001b[0m     layer_outputs \u001b[38;5;241m=\u001b[39m \u001b[43mdecoder_layer\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1019\u001b[0m \u001b[43m        \u001b[49m\u001b[43mhidden_states\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1020\u001b[0m \u001b[43m        \u001b[49m\u001b[43mattention_mask\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcausal_mask\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1021\u001b[0m \u001b[43m        \u001b[49m\u001b[43mposition_ids\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mposition_ids\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1022\u001b[0m \u001b[43m        \u001b[49m\u001b[43mpast_key_value\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mpast_key_values\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1023\u001b[0m \u001b[43m        \u001b[49m\u001b[43moutput_attentions\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43moutput_attentions\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1024\u001b[0m \u001b[43m        \u001b[49m\u001b[43muse_cache\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43muse_cache\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1025\u001b[0m \u001b[43m        \u001b[49m\u001b[43mcache_position\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcache_position\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1026\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1028\u001b[0m hidden_states \u001b[38;5;241m=\u001b[39m layer_outputs[\u001b[38;5;241m0\u001b[39m]\n\u001b[1;32m   1030\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m use_cache:\n",
      "File \u001b[0;32m~/miniconda3/lib/python3.10/site-packages/torch/nn/modules/module.py:1501\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1496\u001b[0m \u001b[38;5;66;03m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1497\u001b[0m \u001b[38;5;66;03m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1498\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m   1499\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1500\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mforward_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1502\u001b[0m \u001b[38;5;66;03m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m   1503\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[38;5;241m=\u001b[39m [], []\n",
      "File \u001b[0;32m~/miniconda3/lib/python3.10/site-packages/transformers/models/llama/modeling_llama.py:741\u001b[0m, in \u001b[0;36mLlamaDecoderLayer.forward\u001b[0;34m(self, hidden_states, attention_mask, position_ids, past_key_value, output_attentions, use_cache, cache_position, **kwargs)\u001b[0m\n\u001b[1;32m    738\u001b[0m hidden_states \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39minput_layernorm(hidden_states)\n\u001b[1;32m    740\u001b[0m \u001b[38;5;66;03m# Self Attention\u001b[39;00m\n\u001b[0;32m--> 741\u001b[0m hidden_states, self_attn_weights, present_key_value \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mself_attn\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    742\u001b[0m \u001b[43m    \u001b[49m\u001b[43mhidden_states\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mhidden_states\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    743\u001b[0m \u001b[43m    \u001b[49m\u001b[43mattention_mask\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mattention_mask\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    744\u001b[0m \u001b[43m    \u001b[49m\u001b[43mposition_ids\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mposition_ids\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    745\u001b[0m \u001b[43m    \u001b[49m\u001b[43mpast_key_value\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mpast_key_value\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    746\u001b[0m \u001b[43m    \u001b[49m\u001b[43moutput_attentions\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43moutput_attentions\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    747\u001b[0m \u001b[43m    \u001b[49m\u001b[43muse_cache\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43muse_cache\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    748\u001b[0m \u001b[43m    \u001b[49m\u001b[43mcache_position\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcache_position\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    749\u001b[0m \u001b[43m    \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    750\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    751\u001b[0m hidden_states \u001b[38;5;241m=\u001b[39m residual \u001b[38;5;241m+\u001b[39m hidden_states\n\u001b[1;32m    753\u001b[0m \u001b[38;5;66;03m# Fully Connected\u001b[39;00m\n",
      "File \u001b[0;32m~/miniconda3/lib/python3.10/site-packages/torch/nn/modules/module.py:1501\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1496\u001b[0m \u001b[38;5;66;03m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1497\u001b[0m \u001b[38;5;66;03m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1498\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m   1499\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1500\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mforward_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1502\u001b[0m \u001b[38;5;66;03m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m   1503\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[38;5;241m=\u001b[39m [], []\n",
      "File \u001b[0;32m~/miniconda3/lib/python3.10/site-packages/transformers/models/llama/modeling_llama.py:360\u001b[0m, in \u001b[0;36mLlamaAttention.forward\u001b[0;34m(self, hidden_states, attention_mask, position_ids, past_key_value, output_attentions, use_cache, cache_position, **kwargs)\u001b[0m\n\u001b[1;32m    357\u001b[0m value_states \u001b[38;5;241m=\u001b[39m value_states\u001b[38;5;241m.\u001b[39mview(bsz, q_len, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mnum_key_value_heads, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mhead_dim)\u001b[38;5;241m.\u001b[39mtranspose(\u001b[38;5;241m1\u001b[39m, \u001b[38;5;241m2\u001b[39m)\n\u001b[1;32m    359\u001b[0m past_key_value \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mgetattr\u001b[39m(\u001b[38;5;28mself\u001b[39m, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mpast_key_value\u001b[39m\u001b[38;5;124m\"\u001b[39m, past_key_value)\n\u001b[0;32m--> 360\u001b[0m cos, sin \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrotary_emb\u001b[49m\u001b[43m(\u001b[49m\u001b[43mvalue_states\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mposition_ids\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    361\u001b[0m query_states, key_states \u001b[38;5;241m=\u001b[39m apply_rotary_pos_emb(query_states, key_states, cos, sin)\n\u001b[1;32m    363\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m past_key_value \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m    364\u001b[0m     \u001b[38;5;66;03m# sin and cos are specific to RoPE models; cache_position needed for the static cache\u001b[39;00m\n",
      "File \u001b[0;32m~/miniconda3/lib/python3.10/site-packages/torch/nn/modules/module.py:1501\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1496\u001b[0m \u001b[38;5;66;03m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1497\u001b[0m \u001b[38;5;66;03m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1498\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m   1499\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1500\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mforward_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1502\u001b[0m \u001b[38;5;66;03m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m   1503\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[38;5;241m=\u001b[39m [], []\n",
      "File \u001b[0;32m~/miniconda3/lib/python3.10/site-packages/torch/utils/_contextlib.py:112\u001b[0m, in \u001b[0;36mcontext_decorator.<locals>.decorate_context\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    109\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m inspect\u001b[38;5;241m.\u001b[39misgeneratorfunction(func):\n\u001b[1;32m    110\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m _wrap_generator(ctx_factory, func)\n\u001b[0;32m--> 112\u001b[0m \u001b[38;5;129m@functools\u001b[39m\u001b[38;5;241m.\u001b[39mwraps(func)\n\u001b[1;32m    113\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mdecorate_context\u001b[39m(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs):\n\u001b[1;32m    114\u001b[0m     \u001b[38;5;28;01mwith\u001b[39;00m ctx_factory():\n\u001b[1;32m    115\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m func(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    },
    {
     "ename": "",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31mThe Kernel crashed while executing code in the the current cell or a previous cell. Please review the code in the cell(s) to identify a possible cause of the failure. Click <a href='https://aka.ms/vscodeJupyterKernelCrash'>here</a> for more info. View Jupyter <a href='command:jupyter.viewOutput'>log</a> for further details."
     ]
    }
   ],
   "source": [
    "if __name__ == \"__main__\":\n",
    "    main()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "During the first part of the journey he travels west for 5 hours at a speed of 18mph so he covered a distance of 5*18 = <<5*18=90>>90 miles\n",
      "Then he travels east at the same speed for 8 hours so he covered a distance of 8*18 = <<8*18=144>>144 miles\n",
      "His distance from his house now is the difference between his east and west movements which is 144-90 = <<144-90=54>>54 miles\n",
      "#### 54\n",
      "I'm ready!\n",
      "\n",
      "Elvis drives west for 5 hours at an average speed of 18 mph, covering a distance of:\n",
      "Distance = Speed × Time = 18 mph × 5 h = 90 miles\n",
      "\n",
      "Then, he turns around and drives east for 8 hours at the same average speed of 18 mph, covering another distance of:\n",
      "Distance = Speed × Time = 18 mph × 8 h = 144 miles\n",
      "\n",
      "To find out how far he is from his house now, we add the distances covered during these two legs of his journey:\n",
      "Total Distance = Westward Journey + Eastward Journey = 90 miles + (-144 miles) = -54 miles\n",
      "\n",
      "Wait, why did I subtract? Because when you travel east after traveling west, you're essentially canceling out some part of your original journey. So, the correct calculation should be:\n",
      "Total Distance = Westward Journey + Eastward Journey = 90 miles + 144 miles = 234 miles\n",
      "\n",
      "However, since Elvis started from his house, we want to know how far away he is from home now. To do that, we'll take the absolute value of our result:\n",
      "Final Distance = | Total Distance | = |-234 miles| = 234 miles\n",
      "\n",
      "Therefore, Elvis is\n"
     ]
    },
    {
     "ename": "",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31mThe Kernel crashed while executing code in the the current cell or a previous cell. Please review the code in the cell(s) to identify a possible cause of the failure. Click <a href='https://aka.ms/vscodeJupyterKernelCrash'>here</a> for more info. View Jupyter <a href='command:jupyter.viewOutput'>log</a> for further details."
     ]
    }
   ],
   "source": [
    "# a = pd.read_csv(os.path.join(args.save_result_dir, f\"result.csv\")).loc[:,[\"mode_answer\",\"model_response\",'ACC']]\n",
    "\n",
    "\n",
    "# aa = a[a['ACC']==False]\n",
    "\n",
    "# print(aa.iloc[388,0])\n",
    "# print(aa.iloc[388,1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['<|begin_of_text|><|start_header_id|>system<|end_header_id|>\\n\\nno<|eot_id|><|start_header_id|>user<|end_header_id|>\\n\\nhello<|eot_id|><|start_header_id|>assistant<|end_header_id|>\\n\\n',\n",
       " '<|begin_of_text|><|start_header_id|>user<|end_header_id|>\\n\\nhi, help<|eot_id|><|start_header_id|>assistant<|end_header_id|>\\n\\n']"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "ename": "",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31mThe Kernel crashed while executing code in the the current cell or a previous cell. Please review the code in the cell(s) to identify a possible cause of the failure. Click <a href='https://aka.ms/vscodeJupyterKernelCrash'>here</a> for more info. View Jupyter <a href='command:jupyter.viewOutput'>log</a> for further details."
     ]
    }
   ],
   "source": [
    "# chat_tokens = tokenizer.apply_chat_template([[{'content': \"no\", \"role\": \"system\"},{'content': \"hello\", 'role': 'user'}],[{'content': \"hi, help\", 'role': 'user'}]], tokenize=False, add_generation_prompt=True)\n",
    "# chat_tokens\n",
    "# ans = model.generate(chat_tokens, eos_token_id = 128009, pad_token_id = tokenizer.pad_token_id)\n",
    "# tokenizer.decode(ans, skip_special_tokens=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[\"Janet’s ducks lay 16 eggs per day. She eats three for breakfast every morning and bakes muffins for her friends every day with four. She sells the remainder at the farmers' market daily for $2 per fresh duck egg. How much in dollars does she make every day at the farmers' market?\",\n",
       " 'A robe takes 2 bolts of blue fiber and half that much white fiber.  How many bolts in total does it take?']"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# dataset['test']['question'][:2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DatasetDict({\n",
       "    train: Dataset({\n",
       "        features: ['question', 'answer'],\n",
       "        num_rows: 7473\n",
       "    })\n",
       "    test: Dataset({\n",
       "        features: ['question', 'answer'],\n",
       "        num_rows: 1319\n",
       "    })\n",
       "})"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'8400'"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# aaa = '''I'll help you solve the problem!\n",
    "\n",
    "# Job A:\n",
    "# Nick earns $15/hour for 2000 hours/year, so his gross income is:\n",
    "# $15/hour × 2000 hours/year = $30,000/year\n",
    "\n",
    "# After deducting 20% taxes, his take-home pay would be:\n",
    "# $30,000/year × (1 - 0.20) = $24,000/year\n",
    "\n",
    "# Job B:\n",
    "# Gross income is fixed at $42,000/year. Subtracting $6000 property tax, we get:\n",
    "# $42,000/year - $6000 = $36,000/year\n",
    "\n",
    "# Net income after paying 10% tax rate on the remaining amount:\n",
    "# $36,000/year × (1 - 0.10) = $32,400/year\n",
    "\n",
    "# Now, let's compare the take-home pay of both jobs:\n",
    "# Job A: $24,000/year\n",
    "# Job B: $32,400/year\n",
    "\n",
    "# To find out how much more Nick will earn at Job B, subtract the take-home pay of Job A from Job B:\n",
    "# $32,400/year - $24,000/year = $8400/year\n",
    "\n",
    "# Therefore, Nick will make $8400 more at Job B compared to Job A.\n",
    "\n",
    "# Answer: The answer is \n",
    "# '''\n",
    "# extract_answer_item(aaa)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}